【摘要】 针对传统可见光定位方法存在空间布局复杂、硬件成本高的局限性,本文提出了一种基于接收信号强度差值的自适应可见光定位虚拟仿真系统。通过改进朗伯辐射模型的光线信道增益公式,推导出光强与距离的非线性映射关系,并创新地融合了局部均值分解降噪算法与自适应扩展卡尔曼滤波算法。仿真结果表明,在0.7m定位范围内,系统平均定位误差为1.038cm,较文献[10]采用的三边测量法提升精度60.1%。实验的创新性在于:通过算法优化将传统三边测量法所需的三传感器方案简化为双传感器与单舵机的组合架构,不仅降低了对多节点空间布局的依赖,还为小范围高精度无源定位提供了更具成本效益的技术解决方案。更多还原

【Abstract】 Aiming at the limitations of traditional visible light positioning methods, such as complex spatial layout and high hardware costs, this paper proposes an adaptive visible light positioning virtual simulation system based on the difference in received signal strength. By improving the Lambertian radiation model’s optical channel gain formula, a nonlinear mapping relationship between light intensity and distance is derived. The system innovatively integrates the local mean decomposition denoising algorithm with the adaptive extended Kalman filter algorithm. Simulation results show that within a positioning range of 0.7m, the system’s average positioning error is 1.038 cm, which is 60.1% more accurate than the trilateration method used in reference [10]. The innovation of the experiment lies in the fact that through algorithm optimization, the traditional trilateration method’s three-sensor scheme is simplified to a combination of two sensors and a single servo. This not only reduces the dependence on multi-node spatial layout but also provides a more cost-effective technical solution for small-range high-precision passive positioning.更多还原